Wan2VideoEditApi - ComfyUI Built-in Node Documentation - ComfyUI

The Wan 2.7 Video Edit node edits a video using text instructions, reference images, or style transfer. It sends the input video (plus any reference images) to the Wan 2.7 video-edit service and returns a newly generated video based on the chosen resolution, aspect ratio, and duration settings.

Inputs

Common Inputs

Parameter	Description	Data Type	Required	Range
`model`	The model to use for video editing. Each option exposes its own set of sub-parameters.	DYNAMIC_COMBO	Yes	`"wan2.7-videoedit"`
`video`	The video to edit.	VIDEO	Yes	-
`seed`	Seed to use for generation. (default: 0)	INT	Yes	0 to 2147483647
`audio_setting`	’auto’: model decides whether to regenerate audio based on the prompt. ‘origin’: preserve the original audio from the input video. (default: “auto”)	COMBO	Yes	`"auto"` `"origin"`
`watermark`	Whether to add an AI-generated watermark to the result. (default: False)	BOOLEAN	Yes	-

wan2.7-videoedit Inputs

Parameter	Description	Data Type	Required	Range
`prompt`	Editing instructions or style transfer requirements. (default: empty string)	STRING	Yes	-
`resolution`	The resolution for the output video.	COMBO	Yes	`"720P"` `"1080P"`
`ratio`	Aspect ratio. If not changed, approximates the input video ratio.	COMBO	Yes	`"16:9"` `"9:16"` `"1:1"` `"4:3"` `"3:4"`
`duration`	Output duration in seconds. ‘auto’ matches the input video duration. A specific value truncates from the start of the video. (default: “auto”)	COMBO	Yes	`"auto"` `"2"` `"3"` `"4"` `"5"` `"6"` `"7"` `"8"` `"9"` `"10"`

Reference Inputs

Parameter	Description	Data Type	Required	Range
`reference_images`	Growable slot: connect 0 to 4 images (`image1`…`image4`) to guide the edit. The count limit is 4 for the wan2.7-videoedit model.	IMAGE	No	0 to 4 items

Constraints:

audio_setting and watermark are advanced options.
The prompt must contain at least 1 character.
The input video must be between 2 and 10 seconds in duration.
The reference_images growable slot accepts a maximum of 4 images.

Outputs

Output Name	Description	Data Type
`output`	The edited video generated by the model.	VIDEO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 27283273ee56c90903db103a3e9bc17dc4df0914676c9aedd2a115b07937dc10